A kind of Fuzzy Controller Based on a Visual Sensor
In the industrial control under worse environment, such as the control of billet in the heating kiln with high temperature, visual sensors are often chosen as a kind of un-contact sensor with excellent property since traditional sensors can hardly detect the useful information. In many cases, the control process is implemented by combining visual sensor (the visual information from videos by a CCD camera) with human-manipulating. But this control method brings out some problems: on the one hand, it requires some special workers in the position to fulfill the control work every day; on the other hand, long time observing the serial videos will cause some harm to the eyes and brains of the workers. Thus, a kind of Fuzzy Controller based on a Visual Sensors is necessary. The information about control targets is obtained by a visual sensor. Based on Fuzzy Logic, the successful manipulating experience of the workers is abstracted as Fuzzy rules and therefore a kind of Fuzzy Controller is designed out. The information of targets from the visual sensor is regarded as the input of the Fuzzy Controller, and the Controller outputs the control information by Fuzzy evaluation. In order to guarantee the real time control, the Fuzzy Controller is implemented by the hardware platform of FPGA. In the application, real time visual feedback makes the Fuzzy Controller control the target nearly as correctly as the skilled workers do. Consequently, some labor is saved and harm for worker is also avoided by adopting the Fuzzy Controller based on a visual sensor.
fuzzy control visual sensor computer vision pattern recognition
Xinhai Liu Kangling Fang Xiaoyu Liu Yan Zhu
College of Information Science & Engineering, Wuhan University of Science and Technology, Wuhan, Chi College of Electric Engineering, Wuhan University, Wuhan, China
国际会议
2006现代科技国际研讨会(The International Workshop on Modern Science and Technology in 2006)
北京
英文
431-435
2006-04-01（万方平台首次上网日期，不代表论文的发表时间）